Kargnas Laravel AI Translator

https://github.com/kargnas/laravel-ai-translator
Admin
AI-powered translation tool for Laravel language files that automates the tedious task of...
Tokens:121,157
Snippets:814
Trust Score:8.5
License:MIT
Update:1 month ago
Show doc for...
Tokens:
Raw
### Using AIProvider translate()

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/_autodocs/api-reference/AIProvider.md

Example of instantiating AIProvider and calling translate() to get an array of LocalizedString items, then iterating to log each key and translated value.

```php
$provider = new AIProvider('auth.php', ['login' => 'Login', 'welcome' => 'Welcome, :name!'], 'en', 'ko');
$items = $provider->translate(); // array<int, LocalizedString>
foreach ($items as $item) {
    Log::info("{$item->key} => {$item->translated}");
}
```

--------------------------------

### Install and run ai-translator:translate

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/_autodocs/api-reference/ServiceProvider.md

Installation and usage flow: install the package, set the OpenRouter API key, optionally publish the config, then run the translate command.

```bash
composer require kargnas/laravel-ai-translator
export OPENROUTER_API_KEY=your-openrouter-api-key-here
php artisan vendor:publish --provider="Kargnas\LaravelAiTranslator\ServiceProvider"  # optional
php artisan ai-translator:translate
```

--------------------------------

### Install @openrouter/ai-sdk-provider

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs-ai/api-openrouter-251016/docs-community-vercel-ai-sdk.txt

Install the OpenRouter provider for the Vercel AI SDK before using streamText() with OpenRouter.

```bash
npm install @openrouter/ai-sdk-provider
```

--------------------------------

### Install Guzzle via Composer

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs/llms-guzzle.txt

Install Guzzle using Composer, the recommended installation method.

```APIDOC
## Install Guzzle via Composer

### Description
Install Guzzle using Composer, the recommended installation method.

### Command
composer require guzzlehttp/guzzle

### Example
```bash
composer require guzzlehttp/guzzle
```
```

--------------------------------

### Create a new Laravel Project and Install Prism

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs-ai/llms-prism.md

Creates a new Laravel project named prism-workshop, installs the Prism PHP library, and publishes Prism's configuration file.

```bash
composer create-project laravel/laravel prism-workshop
cd prism-workshop
composer require prism-php/prism
php artisan vendor:publish --tag=prism-config
```

--------------------------------

### Installing Guzzle via Composer

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs/llms-guzzle.txt

Installs Guzzle via Composer, the recommended installation method.

```bash
composer require guzzlehttp/guzzle
```

--------------------------------

### Initialize Crowdin client and handle errors

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/_autodocs/api-reference/CrowdinServices.md

Example usage of CrowdinApiService: instantiate with a token, call initialize(), catch RuntimeException, and get the client. The initialize() method builds a Crowdin client and verifies connectivity; on failure it throws a RuntimeException with a message that may include a hint about checking API token and organization settings.

```php
$api = new CrowdinApiService($this, $token, null);
try {
    $api->initialize();
} catch (\RuntimeException $e) {
    $this->error($e->getMessage());
    return 1;
}
$crowdin = $api->getClient();
```

--------------------------------

### Basic Guzzle HTTP Client Usage with Authentication

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs/llms-guzzle.txt

Demonstrates how to use Guzzle to make HTTP requests in PHP. The example shows making a synchronous GET request with basic authentication, retrieving response details, and executing an asynchronous request with a promise-based callback.

```APIDOC
## Basic Guzzle HTTP Client Usage with Authentication

### Description
Demonstrates how to use Guzzle to make HTTP requests in PHP. The example shows making a synchronous GET request with basic authentication, retrieving response details, and executing an asynchronous request with a promise-based callback.

### Methods
- **request($method, $uri, array $options = [])** - Sends a synchronous request.
- **sendAsync($request, array $options = [])** - Sends an asynchronous request.

### Request Options
- **auth** (array) - Optional - Basic authentication credentials as [username, password].

### Response Methods
- **getStatusCode()** - Returns the HTTP status code.
- **getHeader($name)** - Returns an array of header values.
- **getBody()** - Returns the response body as a stream.

### Example
```php
$client = new GuzzleHttp\Client();
$res = $client->request('GET', 'https://api.github.com/user', [
    'auth' => ['user', 'pass']
]);
echo $res->getStatusCode();
// "200"
echo $res->getHeader('content-type')[0];
// 'application/json; charset=utf8'
echo $res->getBody();
// {"type":"User"...';

// Send an asynchronous request.
$request = new \GuzzleHttp\Psr7\Request('GET', 'http://httpbin.org');
$promise = $client->sendAsync($request)->then(function ($response) {
    echo 'I completed! ' . $response->getBody();
});
$promise->wait();
```
```

--------------------------------

### Example full configuration

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/README.md

Example full configuration file for config/ai-translator.php. Shows all available settings: source_directory, ai provider/model, locale_names, disable_plural, and additional_rules for customizing translation style. Remember to set your chosen AI provider's API key in .env.

```php
<?php

return [
    'source_directory' => 'lang',

    'ai' => [
        'provider' => 'openrouter',
        'model' => 'vendor/model-id',
        'api_key' => env('OPENROUTER_API_KEY'),
    ],

    'locale_names' => [
        'en' => 'English',
        'ko' => 'Korean',
        'zh_cn' => 'Chinese (Simplified)',
        // ... other locales
    ],

    'disable_plural' => false,

    'additional_rules' => [
        'default' => [
            "Use a friendly and intuitive tone of voice, like the service tone of voice of 'Discord'.",
        ],
        'ko' => [
            "한국의 인터넷 서비스 '토스'의 서비스 말투 처럼, 유저에게 친근하고 직관적인 말투로 설명하고 존댓말로 설명하세요.",
        ],
    ],
];
```

--------------------------------

### TokenChunker usage example

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/_autodocs/api-reference/TranslationHelpers.md

Shows how to instantiate TokenChunker and chunk a source string list with a 1500-token budget, then iterate over the resulting chunks for translation.

```php
$chunks = (new TokenChunker)->chunk($sourceStringList, 1500);
foreach ($chunks as $chunk) { /* translate chunk */ }
```

--------------------------------

### Install package with Composer

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/README.md

Install the package via Composer. Requires PHP 8.3+ and Laravel 12.0+.

```bash
composer require kargnas/laravel-ai-translator
```

--------------------------------

### Stream text with OpenRouter using streamText()

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs-ai/api-openrouter-251016/docs-community-vercel-ai-sdk.txt

Use createOpenRouter() and streamText() to stream text from OpenRouter. Includes a basic prompt example and a tool-calling example with a mock weather function.

```typescript
import { createOpenRouter } from '@openrouter/ai-sdk-provider';
import { streamText } from 'ai';
import { z } from '@openrouter-monorepo/type-utils/zod';

export const getLasagnaRecipe = async (modelName: string) => {
  const openrouter = createOpenRouter({
    apiKey: '${API_KEY_REF}',
  });

  const response = streamText({
    model: openrouter(modelName),
    prompt: 'Write a vegetarian lasagna recipe for 4 people.',
  });

  await response.consumeStream();
  return response.text;
};

export const getWeather = async (modelName: string) => {
  const openrouter = createOpenRouter({
    apiKey: '${API_KEY_REF}',
  });

  const response = streamText({
    model: openrouter(modelName),
    prompt: 'What is the weather in San Francisco, CA in Fahrenheit?',
    tools: {
      getCurrentWeather: {
        description: 'Get the current weather in a given location',
        parameters: z.object({
          location: z
            .string()
            .describe('The city and state, e.g. San Francisco, CA'),
          unit: z.enum(['celsius', 'fahrenheit']).optional(),
        }),
        execute: async ({ location, unit = 'celsius' }) => {
          // Mock response for the weather
          const weatherData = {
            'Boston, MA': {
              celsius: '15°C',
              fahrenheit: '59°F',
            },
            'San Francisco, CA': {
              celsius: '18°C',
              fahrenheit: '64°F',
            },
          };

          const weather = weatherData[location];
          if (!weather) {
            return `Weather data for ${location} is not available.`;
          }

          return `The current weather in ${location} is ${weather[unit]}.`;
        },
      },
    },
  });

  await response.consumeStream();
  return response.text;
};
```

--------------------------------

### OpenRouter assistant prefill with partial assistant message

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs-ai/api-openrouter-251016/docs-api-reference-overview.txt

Use this TypeScript fetch call to prefill an assistant response by including a message with role 'assistant' at the end of the messages array. This guides the model to complete the partial response. The example uses model 'openai/gpt-4o' and includes the required Authorization header.

```typescript
fetch('https://openrouter.ai/api/v1/chat/completions', {
  method: 'POST',
  headers: {
    Authorization: 'Bearer <OPENROUTER_API_KEY>',
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    model: 'openai/gpt-4o',
    messages: [
      { role: 'user', content: 'What is the meaning of life?' },
      { role: 'assistant', content: "I'm not sure, but my best guess is" },
    ],
  }),
});
```

--------------------------------

### Prism Environment Variables for Server and Provider

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs-ai/llms-prism.md

Example .env variables for enabling Prism Server and configuring a provider with an API key and custom endpoint URL.

```shell
# Prism Server Configuration
PRISM_SERVER_ENABLED=true

# Provider Configuration
PROVIDER_API_KEY=your-api-key-here
PROVIDER_URL=https://custom-endpoint.com
```

--------------------------------

### Language::fromCode usage example

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/_autodocs/api-reference/Language.md

Shows how to create a Language instance from a locale code and access its properties. Outputs the name, base code, and plural status for 'ko'.

```php
$lang = Language::fromCode('ko');
echo $lang->name;          // Korean
echo $lang->getBaseCode(); // ko
echo $lang->hasPlural();   // false (ko => 1 form)
```

--------------------------------

### Install Prism PHP Library via Composer

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs-ai/llms-prism.md

Use this command to install the Prism PHP library and its dependencies via Composer. It is advised to pin the version to prevent issues from future breaking changes.

```bash
composer require prism-php/prism
```

--------------------------------

### Usage examples of Utility::getPluralForms

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/_autodocs/api-reference/Utility.md

Shows example calls to Utility::getPluralForms with different locales, demonstrating the return values: 3 for 'ru', 6 for 'ar_eg', 1 for 'ja_jp', and null for 'zz'.

```php
$forms = Utility::getPluralForms('ru');     // 3
$forms = Utility::getPluralForms('ar_eg');  // 6
$forms = Utility::getPluralForms('ja_jp');  // 1
$forms = Utility::getPluralForms('zz');     // null
```

--------------------------------

### English language file example

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/README.md

An example English language file that the package translates. It contains a nested array structure with 'expertise' and 'coding' keys, including a 'title' and a 'description' field.

```php
<?php
return [
    'expertise' => [
        'coding' => [
            'title' => 'Coding & Product',
            'description' => 'Coding is practically my entire life. I started creating fan sites for Harry Potter and StarCraft when I was 11 years old. Before becoming a university student, I developed various game-related web services such as Nexus TK, TalesWeaver, and MapleStory. During my university days, I earned money by doing part-time jobs creating websites. My current company, OP.GG, is also an extension of the game-related services I\'ve been developing since childhood, which gives me a great sense of pride.',
        ],
    ],
];
```

--------------------------------

### setOnThinkingStart(?callable $callback): self

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/_autodocs/api-reference/AIProvider.md

Registers a callback to be called when reasoning starts.

```APIDOC
## setOnThinkingStart(?callable $callback): self

### Description
Called with no arguments when a `ReasoningStart` event arrives.

### Method
`setOnThinkingStart`

### Signature
```php
public function setOnThinkingStart(?callable $callback): self
```

### Parameters
- `$callback` (?callable) - Required - The callback to invoke with no arguments.

### Returns
- `self` - Returns the instance for chaining.
```

--------------------------------

### JSON language file structure

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/README.md

Example directory structure for JSON language files, showing root-level files per locale.

```text
lang/
├── en.json
├── ko.json
├── ja.json
└── fr.json
```

--------------------------------

### Usage examples for ai-translator:find-unused

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/_autodocs/api-reference/CommandsMaintenance.md

Shows three example invocations of the command: default usage, with custom scan paths and JSON output, and with a different source locale and force deletion.

```bash
php artisan ai-translator:find-unused
php artisan ai-translator:find-unused --scan-path=app --scan-path=resources --format=json --show-files
php artisan ai-translator:find-unused --source=ko --force
```

--------------------------------

### find-unused command usage examples

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/README.md

Examples of using the find-unused command with various options: interactive deletion, custom scan paths, force deletion, file details, and JSON output. Backups are created in lang/backup-before-unused/ before deletion.

```bash
# Find unused translations (interactive deletion prompt)
php artisan ai-translator:find-unused

# Scan specific directories
php artisan ai-translator:find-unused --scan-path=app --scan-path=resources

# Auto-delete without confirmation
php artisan ai-translator:find-unused --force

# Show detailed file information
php artisan ai-translator:find-unused --show-files

# Output as JSON
php artisan ai-translator:find-unused --format=json
```

--------------------------------

### Custom Headers Configuration

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs/llms-guzzle.txt

Examples of setting custom HTTP headers for requests, including default headers when creating the client and request-specific headers.

```APIDOC
## Custom Headers Configuration

### Description
Examples of setting custom HTTP headers for requests, including default headers when creating the client and request-specific headers.

### Request Options
- **headers** (array) - Optional - Associative array of headers to set on the request.

### Example
```php
// Set default headers when creating the client
$client = new Client([
    'headers' => [
        'User-Agent' => 'testing/1.0',
        'Accept'     => 'application/json',
    ]
]);

// Set request-specific headers
$response = $client->get('http://httpbin.org/get', [
    'headers' => [
        'X-Foo' => 'Bar',
    ]
]);
```
```

--------------------------------

### PHP language file structure

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/README.md

Example directory structure for PHP language files, showing subdirectories per locale with common files like auth.php, validation.php, and messages.php.

```text
lang/
├── en/
│   ├── auth.php
│   ├── validation.php
│   └── messages.php
├── ko/
│   ├── auth.php
│   ├── validation.php
│   └── messages.php
```

--------------------------------

### Translate a PHP file with ai-translator:translate-file

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/_autodocs/api-reference/CommandsTranslateSingle.md

Example command to translate lang/en/auth.php to Korean with a max context of 200 items. The command writes the translated file to lang/en/auth-ko.php.

```bash
php artisan ai-translator:translate-file lang/en/auth.php --target-language=ko --max-context-items=200
# writes lang/en/auth-ko.php
```

--------------------------------

### Validator usage example

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/_autodocs/api-reference/TranslationHelpers.md

Demonstrates validating a translation with the Validator and logging warnings if issues are found.

```php
$v = new Validator;
$issues = $v->validate('Welcome, :name! <b>Items</b>: {1} item|[2,*] :count items', '환영합니다, :name! <b>항목</b>: {1} 개|[2,*] :count개');
if ($issues !== []) {
    Log::warning(implode('; ', $issues));
}
```

--------------------------------

### Streaming Response with Guzzle in PHP

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs/llms-guzzle.txt

Demonstrates how to stream a response using the stream option. The example shows reading bytes from the response body in chunks until the end of stream is reached.

```php
$response = $client->request('GET', '/stream/20', ['stream' => true]);
// Read bytes off of the stream until the end of the stream is reached
$body = $response->getBody();
while (!$body->eof()) {
    echo $body->read(1024);
}
```

--------------------------------

### Handling Redirects in Guzzle - PHP

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs/llms-guzzle.txt

Examples of handling HTTP redirects with different configurations including disable redirects, default redirects, and custom redirect settings.

```php
$res = $client->request('GET', '/redirect/3', ['allow_redirects' => false]);
echo $res->getStatusCode();
// 302
```

```php
$res = $client->request('GET', '/redirect/3');
echo $res->getStatusCode();
// 200
```

```php
use Psr\Http\Message\RequestInterface;
use Psr\Http\Message\ResponseInterface;
use Psr\Http\Message\UriInterface;

$onRedirect = function(
    RequestInterface $request,
    ResponseInterface $response,
    UriInterface $uri
) {
    echo 'Redirecting! ' . $request->getUri() . ' to ' . $uri . "\n";
};

$res = $client->request('GET', '/redirect/3', [
    'allow_redirects' => [
        'max'             => 10,        // allow at most 10 redirects.
        'strict'          => true,      // use "strict" RFC compliant redirects.
        'referer'         => true,      // add a Referer header
        'protocols'       => ['https'], // only allow https URLs
        'on_redirect'     => $onRedirect,
        'track_redirects' => true
    ]
]);

echo $res->getStatusCode();
// 200

echo $res->getHeaderLine('X-Guzzle-Redirect-History');
// http://first-redirect, http://second-redirect, etc...

echo $res->getHeaderLine('X-Guzzle-Redirect-Status-History');
// 301, 302, etc...
```

--------------------------------

### Handle Redirects with allow_redirects Option

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs/llms-guzzle.txt

Shows how to control redirect behavior with the allow_redirects option. The first example follows redirects (default) and returns 200; the second disables redirects and returns 301.

```php
$response = $client->request('GET', 'http://github.com');
echo $response->getStatusCode();
// 200
```

```php
$response = $client->request('GET', 'http://github.com', [
    'allow_redirects' => false
]);
echo $response->getStatusCode();
// 301
```

--------------------------------

### Authentication in Guzzle - PHP

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs/llms-guzzle.txt

Examples of different authentication methods including basic, digest, and NTLM authentication.

```php
$client->request('GET', '/get', ['auth' => ['username', 'password']]);
```

```php
$client->request('GET', '/get', [
    'auth' => ['username', 'password', 'digest']
]);
```

```php
$client->request('GET', '/get', [
    'auth' => ['username', 'password', 'ntlm']
]);
```

--------------------------------

### Configure provider in config/ai-translator.php

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/README.md

Configures the matching provider, model, and API key in config/ai-translator.php as part of the provider setup steps. This is the same structure as the general ai configuration.

```php
'ai' => [
   'provider' => 'openrouter',
   'model' => 'vendor/model-id',
   'api_key' => env('OPENROUTER_API_KEY'),
],
```

--------------------------------

### Create Stream with GuzzleHttp\Psr7\Utils::streamFor

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs/llms-guzzle.txt

Creates a stream from a string using GuzzleHttp\Psr7\Utils::streamFor. The example shows reading from the stream, getting contents, checking if the stream has reached the end, and getting the current position.

```php
use GuzzleHttp\Psr7;

$stream = Psr7\Utils::streamFor('string data');
echo $stream;
// string data
echo $stream->read(3);
// str
echo $stream->getContents();
// ing data
var_export($stream->eof());
// true
var_export($stream->tell());
// 11
```

--------------------------------

### Using ChangeDetector to translate only changed strings

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/_autodocs/api-reference/TranslationHelpers.md

Example usage: instantiate ChangeDetector, get changed strings, translate only those, then save state.

```php
$detector = new ChangeDetector;
$changed = $detector->changedAgainstState($flat, 'en', 'ko');
foreach ($changed as $key => $value) { /* translate only $changed */ }
$detector->saveState(array_keys($changed), $flat, 'en', 'ko');
```

--------------------------------

### JavaScript example for printing errors

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs-ai/api-openrouter-251016/docs-api-reference-errors.txt

Shows how to fetch from OpenRouter and print the HTTP status and error details. The status will be an error code unless the model started processing the request.

```typescript
const request = await fetch('https://openrouter.ai/...');
console.log(request.status); // Will be an error code unless the model started processing your request
const response = await request.json();
console.error(response.error?.status); // Will be an error code
console.error(response.error?.message);
```

--------------------------------

### Create a Basic Search Tool with Prism Tool Facade

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs-ai/llms-prism.md

Defines a 'search' tool with a string parameter 'query' and a closure that returns a string. The tool must return a string.

```php
use Prism\Prism\Facades\Tool;

$searchTool = Tool::as('search')
    ->for('Search for current information')
    ->withStringParameter('query', 'The search query')
    ->using(function (string $query): string {
        // Your search implementation
        return "Search results for: {$query}";
    });
```

--------------------------------

### Creating PSR-7 Request Objects in PHP

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs/llms-guzzle.txt

Shows how to create a new PSR-7 Request object using GuzzleHttp\Psr7\Request class. Includes examples of creating simple GET requests and more complex requests with headers and body content.

```php
use GuzzleHttp\Psr7\Request;

$request = new Request('GET', 'http://httpbin.org/get');

// You can provide other optional constructor arguments.
$headers = ['X-Foo' => 'Bar'];
$body = 'hello!';
$request = new Request('PUT', 'http://httpbin.org/put', $headers, $body);
```

--------------------------------

### Cancel streaming request in Python using requests and threading

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs-ai/api-openrouter-251016/docs-api-reference-streaming.txt

Uses a cancel_event to stop a streaming request. The response is closed when the event is set, stopping model processing and billing for supported providers. Example usage starts a thread and sets the event to cancel.

```python
import requests
from threading import Event, Thread

def stream_with_cancellation(prompt: str, cancel_event: Event):
    with requests.Session() as session:
        response = session.post(
            "https://openrouter.ai/api/v1/chat/completions",
            headers={"Authorization": f"Bearer {{API_KEY_REF}}"},
            json={"model": "{{MODEL}}", "messages": [{"role": "user", "content": prompt}], "stream": True},
            stream=True
        )

        try:
            for line in response.iter_lines():
                if cancel_event.is_set():
                    response.close()
                    return
                if line:
                    print(line.decode(), end="", flush=True)
        finally:
            response.close()

# Example usage:
cancel_event = Event()
stream_thread = Thread(target=lambda: stream_with_cancellation("Write a story", cancel_event))
stream_thread.start()

# To cancel the stream:
cancel_event.set()
```

--------------------------------

### ServiceProvider class with boot and register

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/_autodocs/api-reference/ServiceProvider.md

The full ServiceProvider class. boot() publishes the config file; register() registers the AiServiceProvider dependency, merges the config under the 'ai-translator' namespace, and registers all nine Artisan commands.

```php
class ServiceProvider extends \Illuminate\Support\ServiceProvider
{
    public function boot(): void
    {
        $this->publishes([
            __DIR__.'/../config/ai-translator.php' => config_path('ai-translator.php'),
        ]);
    }

    public function register(): void
    {
        $this->app->register(AiServiceProvider::class);

        $this->mergeConfigFrom(
            __DIR__.'/../config/ai-translator.php',
            'ai-translator',
        );

        $this->commands([
            CleanCommand::class,
            FindUnusedTranslations::class,
            TranslateStrings::class,
            TranslateStringsParallel::class,
            TranslateCrowdinParallel::class,
            TranslateCrowdin::class,
            TestTranslateCommand::class,
            TranslateFileCommand::class,
            TranslateJson::class,
        ]);
    }
}
```

--------------------------------

### Generate Text with Different Providers (Mistral, Ollama, OpenAI)

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs-ai/llms-prism.md

These three examples show how to generate text with Prism using different providers (Mistral, Ollama, and OpenAI) while keeping the same system prompt and user prompt. They demonstrate provider interoperability by using `withSystemPrompt` with a Laravel view for the system prompt.

```php
use Prism\Prism\Prism;
use Prism\Prism\Enums\Provider;

$response = Prism::text()
    ->using(Provider::Mistral, 'mistral-medium')
    ->withSystemPrompt(view('prompts.system'))
    ->withPrompt('Explain quantum computing to a 5-year-old.')
    ->asText();

echo $response->text;
```

```php
use Prism\Prism\Prism;
use Prism\Prism\Enums\Provider;

$response = Prism::text()
    ->using(Provider::Ollama, 'llama2')
    ->withSystemPrompt(view('prompts.system'))
    ->withPrompt('Explain quantum computing to a 5-year-old.')
    ->asText();

echo $response->text;
```

```php
use Prism\Prism\Prism;
use Prism\Prism\Enums\Provider;

$response = Prism::text()
    ->using(Provider::OpenAI, 'gpt-4')
    ->withSystemPrompt(view('prompts.system'))
    ->withPrompt('Explain quantum computing to a 5-year-old.')
    ->asText();

echo $response->text;
```

--------------------------------

### Streaming with reasoning tokens (Python)

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs-ai/api-openrouter-251016/docs-use-cases-reasoning-tokens.txt

Python example using the OpenAI SDK to stream a chat completion with a reasoning token budget of 8000 tokens. The response is streamed and reasoning/content deltas are printed separately. Requires an API key and uses the model 'anthropic/claude-3.7-sonnet'.

```python
from openai import OpenAI

client = OpenAI(
    base_url="https://openrouter.ai/api/v1",
    api_key="{{API_KEY_REF}}",
)

def chat_completion_with_reasoning(messages):
    response = client.chat.completions.create(
        model="{{MODEL}}",
        messages=messages,
        max_tokens=10000,
        reasoning={
            "max_tokens": 8000  # Directly specify reasoning token budget
        },
        stream=True
    )
    return response

for chunk in chat_completion_with_reasoning([
    {"role": "user", "content": "What's bigger, 9.9 or 9.11?"}
]):
    if hasattr(chunk.choices[0].delta, 'reasoning') and chunk.choices[0].delta.reasoning:
        print(f"REASONING: {chunk.choices[0].delta.reasoning}")
    elif chunk.choices[0].delta.content:
        print(f"CONTENT: {chunk.choices[0].delta.content}")
```

--------------------------------

### Japanese translation example

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/README.md

Generated Japanese translation of the English example. The title and description are fully translated.

```php
<?php
return [
    'expertise' => [
        'coding' => [
            'title' => 'コーディング＆プロダクト',
            'description' => 'コーディングは私の人生そのものです。11歳の時からハリーポッターやスタークラフトのファンサイトを作り始めました。大学生になる前に、Nexus TK、テイルズウィーバー、メイプルストーリーなどのゲーム関連のウェブサービスを開発していました。大学時代には、ウェブサイト制作のアルバイトでお金を稼いでいました。現在の会社OP.GGも、子供の頃から開発してきたゲーム関連サービスの延長線上にあり、それが大きな誇りとなっています。',
          ]
      ]
  ];
```

--------------------------------

### Use `withSystemPrompt` for Provider Interoperability

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs-ai/llms-prism.md

Shows the recommended pattern for provider interoperability: avoid using `SystemMessage` directly in `withMessages` when switching providers, and instead use `withSystemPrompt` so Prism can handle provider-specific formatting.

```php
// Avoid this when switching between providers
$response = Prism::text()
    ->using(Provider::OpenAI, 'gpt-4o')
    ->withMessages([
        new SystemMessage('You are a helpful assistant.'),
        new UserMessage('Tell me about AI'),
    ])
    ->asText();

// Prefer this instead
$response = Prism::text()
    ->using(Provider::OpenAI, 'gpt-4o')
    ->withSystemPrompt('You are a helpful assistant.')
    ->withPrompt('Tell me about AI')
    ->asText();
```

--------------------------------

### Thai translation example

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/README.md

Generated Thai translation of the English example. The title remains in English while the description is translated.

```php
<?php
return [
    'expertise' => [
        'coding' => [
            'title' => 'Coding & Product',
            'description' => 'การเขียนโค้ดเป็นเรื่องที่อยู่ในชีวิตผมแทบทั้งหมด ผมเริ่มสร้างเว็บไซต์แฟนคลับสำหรับแฮร์รี่ พอตเตอร์และสตาร์คราฟท์ตั้งแต่อายุ 11 ปี ก่อนที่จะเข้ามหาวิทยาลัย ผมได้พัฒนาเว็บบริการเกี่ยวกับเกมต่างๆ เช่น Nexus TK, TalesWeaver และ MapleStory ในช่วงมหาวิทยาลัย ผมหาเงินด้วยการทำงานพาร์ทไทม์สร้างเว็บไซต์ บริษัทปัจจุบันของผม OP.GG ก็เป็นส่วนขยายของบริการเกี่ยวกับเกมที่ผมได้พัฒนามาตั้งแต่เด็ก ซึ่งทำให้ผมรู้สึกภูมิใจเป็นอย่างมาก',
          ]
      ]
  ];
```

--------------------------------

### Korean translation example

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/README.md

Generated Korean translation of the English example. The structure mirrors the original with translated title and description.

```php
<?php
return [
    'expertise' => [
        'coding' => [
            'title' => '코딩 & 제품',
            'description' => '코딩은 사실상 제 인생 전부입니다. 11살 때부터 해리 포터와 스타크래프트 팬 사이트를 만들기 시작했습니다. 대학생이 되기 전에 넥서스 TK, 테일즈위버, 메이플스토리와 같은 다양한 게임 관련 웹 서비스를 개발했습니다. 대학 시절에는 웹사이트를 만드는 아르바이트로 돈을 벌었습니다. 현재 제 회사인 OP.GG도 어린 시절부터 개발해 온 게임 관련 서비스의 연장선으로, 이는 저에게 큰 자부심을 줍니다.',
          ]
      ]
  ];
```

--------------------------------

### Clean command usage

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/README.md

Shows the basic usage of the clean command with optional pattern and options.

```bash
php artisan ai-translator:clean [pattern] [options]
```

--------------------------------

### English (Reddit) translation example

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/README.md

Generated English (Reddit) translation of the English example. The title remains the same while the description is rewritten in a casual, Reddit-style tone.

```php
<?php
return [
    'expertise' => [
        'coding' => [
            'title' => 'Coding & Product',
            'description' => 'Coding is practically my entire life, duh. Started building Harry Potter and StarCraft fan sites at 11 (yeah, I was *that* kid). Before even hitting university, I was already knee-deep in game sites like Nexus TK, TalesWeaver, and MapleStory. Paid my way through college building websites - because who needs a social life, right? Now I run OP.GG, which is basically just the grown-up version of what little-me was doing in his bedroom. Not gonna lie, pretty damn proud of that full-circle moment.',
          ]
      ]
  ];
```

--------------------------------

### Initializing Guzzle HTTP Client - PHP

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs/llms-guzzle.txt

Creates a new Guzzle HTTP client instance with a base URI configuration.

```php
$client = new GuzzleHttp\Client(['base_uri' => 'http://httpbin.org']);
```

--------------------------------

### Simplified Chinese translation example

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/README.md

Generated Simplified Chinese translation of the English example. Note the extra spaces in the description text as shown in the source.

```php
<?php
return [
    'expertise' => [
        'coding' => [
            'title' => '编程与产品',
            'description' => '编程几乎就是我的整个生活。11岁时，我就开始为《哈利·波特》和《星际争霸》创建粉丝网站。在上大学之前，我开发了各种游戏相关网络服务，如Nexus TK、TalesWeaver和冒险岛。大学期间，我通过创建网站的兼职工作赚钱。我现在的公司OP.GG，也  是我从小就在开发的游戏相关服务的延伸，这让我感到非常自豪。'
          ]
      ]
  ];
```

--------------------------------

### Example JSON response

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs-ai/api-openrouter-251016/docs-api-reference-overview.txt

Example of a non-streaming completion response. The 'message' field is present because streaming is not used. The 'model' field shows the provider/model identifier.

```json
{
  "id": "gen-xxxxxxxxxxxxxx",
  "choices": [
    {
      "finish_reason": "stop", // Normalized finish_reason
      "native_finish_reason": "stop", // The raw finish_reason from the provider
      "message": {
        // will be "delta" if streaming
        "role": "assistant",
        "content": "Hello there!"
      }
    }
  ],
  "usage": {
    "prompt_tokens": 0,
    "completion_tokens": 4,
    "total_tokens": 4
  },
  "model": "openai/gpt-3.5-turbo" // Could also be "anthropic/claude-2.1", etc, depending on the "model" that ends up being used
}
```

--------------------------------

### Basic Guzzle HTTP Client Usage in PHP

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs/llms-guzzle.txt

Shows how to create a Guzzle client, send a synchronous GET request, inspect the status code, header line, and body, and then send an asynchronous request using sendAsync() with a promise callback that must be waited on with wait().

```php
$client = new \GuzzleHttp\Client();
$response = $client->request('GET', 'https://api.github.com/repos/guzzle/guzzle');

echo $response->getStatusCode(); // 200
echo $response->getHeaderLine('content-type'); // 'application/json; charset=utf8'
echo $response->getBody(); // '{"id": 1420053, "name": "guzzle", ...}'

// Send an asynchronous request.
$request = new \GuzzleHttp\Psr7\Request('GET', 'http://httpbin.org');
$promise = $client->sendAsync($request)->then(function ($response) {
    echo 'I completed! ' . $response->getBody();
});

$promise->wait();
```

--------------------------------

### Korean (North Korea) translation example

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/README.md

Generated Korean (North Korea) translation of the English example. The title and description use North Korean vocabulary and style.

```php
<?php
return [
    'expertise' => [
        'coding' => [
            'title' => '콤퓨터프로그람작성 & 제품',
            'description' => '프로그람작성은 혁명적으로 나의 전생애입니다. 11살때부터 해리 포터와 스타크래프트 애호가싸이트를 만들기 시작하였습니다. 대학생이 되기전에 넥서스 TK, 테일즈위버, 메이플스토리와 같은 혁명적인 게임관련 망봉사들을 개발하였습니다. 대학시절에는 웹싸이트를 만드는 림시일로 돈을 벌었습니다. 현재 나의 회사인 OP.GG도 어린시절부터 개발하여온 게임관련 봉사의 련장선으로, 이는 나에게 큰 혁명적자부심을 줍니다.',
          ]
      ]
  ];
```

--------------------------------

### Instantiate and print full report

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/_autodocs/api-reference/TokenUsagePrinter.md

Creates a TokenUsagePrinter with the translator's model and prints the full report including summary, cost estimation, and model comparison.

```php
$printer = new TokenUsagePrinter($translator->getModel());
$printer->printFullReport($this, $translator->getTokenUsage());
```

--------------------------------

### Simplified Chinese (Dongbei Dialect) translation example

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/README.md

Generated Simplified Chinese (Dongbei Dialect) translation of the English example. The title and description use colloquial dialect expressions.

```php
<?php
return [
    'expertise' => [
        'coding' => [
            'title' => '写代码和整产品',
            'description' => '写代码简直就是俺的命根子。俺11岁那会儿就开始整哈利波特和星际争霸的粉丝网站了。上大学前，俺整了不少游戏网站，啥Nexus TK啊，TalesWeaver啊，还有冒险岛啊。上大学那会儿，俺靠给人整网站挣了不少零花钱。现在俺这公司OP.GG，也就是俺从小整的游戏服务的延续，这让俺贼有成就感。'
          ]
      ]
  ];
```

--------------------------------

### Set up Basic Text Response Faking with Prism PHP

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs-ai/llms-prism.md

Set up a basic fake text response using TextResponseFake::make() with expected text and usage, then intercept calls with Prism::fake() and assert the output. The example uses the Anthropic provider with 'claude-3-5-sonnet-latest' and expects the text 'Hello, I am Claude!'.

```php
use Prism\Prism\Prism;
use Prism\Prism\Enums\Provider;
use Prism\Prism\ValueObjects\Usage;
use Prism\Prism\Testing\TextResponseFake;

it('can generate text', function () {
    $fakeResponse = TextResponseFake::make()
        ->withText('Hello, I am Claude!')
        ->withUsage(new Usage(10, 20));

    // Set up the fake
    $fake = Prism::fake([$fakeResponse]);

    // Run your code
    $response = Prism::text()
        ->using(Provider::Anthropic, 'claude-3-5-sonnet-latest')
        ->withPrompt('Who are you?')
        ->asText();

    // Make assertions
    expect($response->text)->toBe('Hello, I am Claude!');
});
```

--------------------------------

### Set AI behavior with system prompts in Prism (generate)

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/docs-ai/llms-prism.md

Alternative example of setting a system prompt for an expert mathematician persona, using the generate() method instead of asText(). Uses Anthropic's claude-3-sonnet model.

```php
$response = Prism::text()
    ->using(Provider::Anthropic, 'claude-3-sonnet')
    ->withSystemPrompt('You are an expert mathematician who explains concepts simply.')
    ->withPrompt('Explain the Pythagorean theorem.')
    ->generate();
```

--------------------------------

### Korean (Busan Dialect) translation example

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/README.md

Generated Korean (Busan Dialect) translation of the English example. The title remains standard Korean while the description uses Busan dialect expressions.

```php
<?php
return [
    'expertise' => [
        'coding' => [
            'title' => '코딩 & 제품',
            'description' => '코딩은 사실상 제 인생 전부라 카모. 11살 때부터 해리 포터캉 스타크래프트 팬 사이트 만들기 시작했데이. 대학생이 되기 전에는 넥서스 TK, 테일즈위버, 메이플스토리 같은 여러 게임 관련 웹 서비스 개발했디. 대학 다닐 때는 웹사이트 만드는 아르바이트로 돈 벌었데이. 지금 제 회사인 OP.GG도 어릴 때부터 개발해 온 게임 관련 서비스으 연장선인데, 이기 내인데 억수로 자부심이 된다 카모.',
          ]
      ]
  ];
```

--------------------------------

### Run OpenRouter smoke test

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/scripts/README-smoke.md

Sets up a disposable Laravel host project, installs the package, modifies the config to use OpenRouter with the specified model and API key, and runs the translation test command. Do not run against production language files.

```bash
export OPENROUTER_API_KEY='your-openrouter-api-key'
./scripts/test-setup.sh laravel-ai-translator-openrouter-smoke
cd laravel-ai-translator-openrouter-smoke

perl -0pi -e "s/'provider' => 'anthropic'/'provider' => 'openrouter'/; s/'model' => 'claude-3-5-sonnet-latest'/'model' => 'anthropic\/claude-sonnet-4.5'/; s/env\('ANTHROPIC_API_KEY'\)/env('OPENROUTER_API_KEY')/" config/ai-translator.php

php artisan ai-translator:test en ko --text='Hello from OpenRouter'
```

--------------------------------

### Incremental parsing with parseChunk()

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/_autodocs/api-reference/AIResponseParser.md

Incremental parsing example. Feed streaming chunks to parseChunk() and retrieve the accumulated items with getTranslatedItems(). The example yields one item with key 'a' and translated '안녕'.

```php
$parser = new AIResponseParser();
$parser->parseChunk('<translations><item><key>a</key><trx><![CDATA[');
$parser->parseChunk('안녕]]></trx></item></translations>');
$items = $parser->getTranslatedItems(); // 1 item: key='a', translated='안녕'
```

--------------------------------

### Example usage of ai-translator:translate-crowdin-parallel

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/_autodocs/api-reference/CommandsCrowdin.md

Example invocation of the parallel translation command with project ID 123, source language English, max processes 5, and chunk size 50.

```bash
php artisan ai-translator:translate-crowdin-parallel --project=123 --source-language=English \
    --max-processes=5 --chunk-size=50
```

--------------------------------

### getModel(): string

Source: https://github.com/kargnas/laravel-ai-translator/blob/master/_autodocs/api-reference/ConsensusTranslator.md

Get the model name used for consensus translation.

```APIDOC
## getModel(): string

### Description
Returns the model name used for the consensus translation. It returns the judge model if set, otherwise the first translator's model, otherwise 'unknown'.

### Method
getModel

### Returns
- **string** - The model name.

```
Tokens:
Raw